Species Composition, Abundance, and Vertical Distribution of the Stomiid (Pisces: Stomiiformes) Fish Assemblage of the Gulf of Mexico
Total Page:16
File Type:pdf, Size:1020Kb
BULLETIN OF MARINE SCIENCE, 59(3): 530-542, 1996 SPECIES COMPOSITION, ABUNDANCE, AND VERTICAL DISTRIBUTION OF THE STOMIID (PISCES: STOMIIFORMES) FISH ASSEMBLAGE OF THE GULF OF MEXICO Tracey T, Sutton and Thomas L. Hopkins ABSTRACT Species composition, abundance, and vertical distribution of the stomiid fish assemblage were investigated in the eastern Gulf of Mexico, a low-latitude, oligotrophic oceanic ecosys- tem. Seventy-two described species, representing 18 genera, and one undescribed species were identified from 1155 trawl samples. With an additional 10 species reported elsewhere, the stomiid species number now known equals 83, making the Stomiidae the most diverse fish family in the Gulf of Mexico. The assemblage was dominated by three species, Pholos- lomias guemei, Chauliodus s/oani and Stomias affinis. These species, as well as four other common species, exhibited an asynchronous diel vertical migration pattern (450-900 m dur- ing day; 20-300, 550-900 m at night). The percentage of the populations of the three dom- inant species migrating daily ranged from 50-70%. Two other patterns occurred in less abundant species: synchronous migration (400-700 m during the day, 0-200 m at night); and, possible migration from the bathypelagial (> 1000 m during day; 50-300 m at night). Minimum abundance and biomass estimates for the entire assemblage were 1.86 X 105 in- dividuals and 35.3 kg DW·km-2 in the upper 1000 m. Stomiids comprised approximately 10% of the micronekton standing stock in the eastern Gulf. Extrapolating eastern Gulf data to the world warm-water mesopelagial, abundance results suggest that stomiids are the dom- inant mesopelagic upper-trophic level predatory fishes, and as such may serve as key trophic mediators in the transfer of energy in these ecosystems. The "black" stomioid fishes of the family Stomiidae, sensu Fink (1985), are among the most specialized for a predatory oceanic existence (Tchernavin, 1953; Marshall, 1954), with characters including elongate black bodies, highly disten- sible stomachs, and mental barbels which often bear elaborate bioluminescent organs. Faunal studies from the western tropical Pacific, Hawaii, and the eastern North Atlantic show that stomiid assemblages are extremely species-rich in warm- water mesope1agic ecosystems (Parin et at, 1977; Clarke, 1982; Craddock et at, 1987). Despite the potential importance of these fishes in midwater food webs, very little is known of their abundance, vertical and geographic distributions, and assemblage structure (Haffner, 1952; Gibbs, 1969; Backus et at, 1970; Clarke, 1974, 1982; Badcock and Merrett, 1977; Blackburn, 1981). In this first paper we describe the structure of the stomiid assemblage found in the upper 1000 m of the eastern Gulf of Mexico. In a subsequent paper we will describe the feeding ecology of the Stomiidae. Hydrographic Setting.-Circulation in the eastern Gulf of Mexico is dominated by the flow of the Subtropical Undercurrent (Florida Loop Current), which enters the Gulf through the Yucatan Straits and exits through the Florida Straits (Leipper, 1970; Nowlin, 1971; Maul, 1977). The extent of intrusion into the Gulf is lati- tudinally and seasonally unpredictable. The sampling site (2rN 86°W), located east of the axis of the Loop Current, is predominantly occupied by residual eastern Gulf water, which can be differentiated from Loop Current water by the depth of the 22°C isotherm (Leipper, 1970; Jones, 1973). The only sampling period in which Loop Current water was known to be present in the area was March 1985. Temperature profiles from the area show a surface mixed layer with a depth 530 SUTTON AND HOPKINS: STOMIJDS OF THE GULF OF MEXICO 531 between 30 m and 50 m and warm-month temperatures between 27°C and 30°C. The thermocline extends from the bottom of the mixed layer to approximately 150 m, with temperatures at the lower depth between 15°C and 18°C. Tempera- tures decrease to about 4°C at 1000 m. The water column is well oxygenated, having a minimum concentration of 2.7 to 2.9 ml/liter between 400 and 500 m (Hopkins, unpubI. data; Nowlin, 1971). Primary productivity measurements indicate that the sampling area lies in oli- gotrophic waters, with an annual production of around 100 gC.y-l (EI-Sayed and Thrner, 1977). Residual eastern Gulf water, though oligotrophic, appears to be somewhat more productive than Loop Current water (Jones, 1973; Hopkins, 1982). It is estimated that mixing between the two water masses is only 10% (Passarella and Hopkins, 1991), allowing the development of a unique faunal community in eastern Gulf residual water (Michel and Foyo, 1976; Gartner et aI., 1987; Passarella and Hopkins, 1991; Flock and Hopkins, 1992; Richards et aI., 1993). The zooplankton standing stock of the upper 1000 m in the area has been estimated to be 1.2 gDW·m-2 (Hopkins, 1982), which is within the range of values found for oligotrophic boundary currents (Vinogradov, 1970). The hydrography of the eastern Gulf of Mexico, then, is characteristic of non-upwelling, seasonally stable, vertically stratified, oligotrophic, subtropical-tropical oceanic waters (McGowan, 1974; Longhurst, 1976). This area should represent a reasonable an- alog to the low-latitude oceanic gyre systems. Trophic organization and energy transfer are thought to be relatively advanced (i.e., high transfer efficiency be- tween trophic levels) in low-latitude gyre systems due to resource limitations and hydrographic stability (Ryther, 1969; Steele, 1974; Sheldon et aI., 1977; Mc- Gowan, 1977), allowing the high diversity seen in tropical midwater faunas in general (Gibbs and Roper, 1970; Badcock, 1970) and in particular in the stomiid assemblage of the eastern Gulf. METHODS Specimens were sorted from 1,176 trawl samples taken on 24 cruises. These cruises occurred over an 19-year span and covered all four seasons (Table 1), though with most collections (71%) taken during the warm months between May and September. Sampling was centered within a 35 Ian radius of 27°N 86"W.This site has a depth greater than 3,000 m and is located far enough from the continental shelf that stragglers from "boundary communities" are rarely encountered (Hopkins et aI., 1981; Reid et aI., 1991; Hulley, 1992). Sampling prior to August 1987 was conducted with mouth opening-closing Tucker trawls of 1.6 mm or 4.0 mm mesh netting and effective mouth areas of 2.6 m2 or 5.3 m2, based on a mouth angle of 35° from vertical (SCUBA observations) measured at a towing speed of 2 kn. The trawls were fitted with 0.33-, 0.5-, or I-mm-mesh cod end plankton nets. Depth was monitored using a depth transducer/conducting cable system, excepting earlier cruises (Mizar I, Bellows I-III) where triangu- lation was used. A time-depth recorder attached to the trawl frame served as a back-up record on all tows. The trawls were opened and closed either by a messenger-operated double release mechanism (Hopkins et aI., 1973) or by paired clock release mechanisms (Davies and Barham, 1969). The volume of water filtered was measured by a dial-type flow meter which recorded only when the trawl was fishing. The filtration efficiency for all volume calculations was assumed to be 100%. Expendable bathythermograph (XBT) andlor CTD casts were made on all cruises to determine the temperature profiles from 0-1,000 m. Sampling after August 1987 was conducted with a seven-net 4-m2 MOCNESS midwater trawl (Wiebe et aI., 1976) towed at 2 to 4 kn. Physical data and volume sampled were recorded during sampling with a CTD and TSK flowmeter, respectively. These data were stored via conducting cable! microprocessor system. Samples were fixed in 10% (v/v) buffered formalin and later transferred to 50% isopropanol or 70% ethanol. Stomiid specimens were identified according to the keys and revisions of Gibbs (1964a, 1964b), Morrow (1964a, 1964b, 1964c), Morrow and Gibbs (1964), Weitzman (1967), Barnett and Gibbs (1968), Goodyear and Gibbs (1969), Gibbs et al. (1983), Gomon and Gibbs (1985) and Fink and Fink (1986). The status of the familial classification of the barbeled stomioids has not reached 532 BULLETINOFMARINESCIENCE.VOL.59. NO.3. 1996 Table I. Sampling data. All tows taken in the vicinity of 2rN 36°W Cruise Date Sampling gear No. tows MIZARI Jun 1971 Tucker trawl 21 BELLOWSI Aug 1972 Thcker trawl 25 BELLOWSII Oct 1973 Thcker trawl 11 BELLOWSIII Aug 1974 Tucker trawl 15 COLUMBUSISELINI Jun 1975 Thcker trawl 29 COLUMBUSISELINII Jun 1976 Tucker trawl 133 BELLOWSIV Jun 1977 Tucker trawl 52 COLUMBUSISELINIII Oct 1977 Thcker trawl 96 BELLOWSVII Jun 1981 Thcker trawl 3 BELLOWSVIII Aug 1981 Tucker trawl 66 BELLOWSIX Jul 1982 Tucker trawl 62 BELLOWSX Aug 1982 Thcker trawl 30 SUNCOASTERI Aug 1984 Tucker trawl 52 SUNCOASTERII Mar 1985 Tucker trawl 66 SUNCOASTERIII Jul 1985 Thcker trawl 59 SUNCOASTERIV Nov 1985 Thcker trawl 49 SUNCOASTERV Jan 1986 Thcker trawl 49 SUNCOASTERVI May 1986 Thcker trawl 48 SUNCOASTERVII Jan 1987 Thcker trawl 34 SUNCOASTERVIII Mar 1987 Thcker trawl 30 SUNCOASTERIX Aug 1987 MOCNESS 55 SUNCOASTERX Sep 1987 MOCNESS 20 SUNCOASTERXI Jul 1989 MOCNESS 99 SUNCOASTERXII Jul 1990 MOCNESS 72 24 1971-1990 1,176 universal agreement; some workers favor the single family (Stomiidae) scheme of Fink (1985) (Gibbs and McKinney, 1988; Nelson, 1994), while others have opted for the six family scheme listed in Weitzman (1974) (Eschmeyer, 1990; Reid et aI., 1991). The single family classification scheme (Fink, 1985) is followed here with acceptance of the subfamilies Astronesthinae, Idiacanthinae, Malacostei- nae, Melanostomiinae, and Stomiinae (Nelson, 1994). Specimens were measured to the nearest millimeter standard length (SL). Corrections for shrinkage due to preservation were not attempted because no fresh specimens were available for comparison. Shrinkage has been estimated to be 5% SL of preserved melanostomiines (Beebe and Crane, 1939) and 12% SL of preserved myctophids (Gartner et aI., 1987).